Résumé : Background: A high prevalence of epilepsy has been observed in several onchocerciasis-endemic countries, including Cameroon. However, little is known on the clinical presentations of the affected persons with epilepsy (PWE). A community-based study was conducted with the aim of describing the spectrum of seizures in selected onchocerciasis-endemic villages in Cameroon and documenting relevant medical history in patients with onchocerciasis-associated epilepsy (OAE). Methods: We carried out door-to-door surveys in 5 onchocerciasis-endemic villages in Cameroon and recruited all consenting PWE. Epilepsy was diagnosed using a 2-step approach consisting of the administration of a standardized 5-item questionnaire followed by confirmation of the suspected cases by a neurologist. Onchocerciasis-associated epilepsy was defined as ≥ 2 seizures without an obvious cause, starting between the ages of 3–18 years in previously healthy persons having resided for at least 3 years in an onchocerciasis-endemic area. Ivermectin use by PWE was verified. Seizure history, relevant past medical, and family history, as well as neurological findings, were noted. Results: In all, 156 PWE were recruited in the 5 villages. The modal age group for epilepsy onset was 10–14 years. The diagnostic criteria for OAE were met by 93.2% of the PWE. Participants had one or more of the following seizure types: generalized tonic–clonic seizures (89.1%), absences (38.5%), nodding (21.8%), focal nonmotor (7.7%), and focal motor seizures (1.9%). One case (0.6%) with the “Nakalanga syndrome” was identified. More than half (56.4%) of PWE had at least one seizure per month. In one village, 56.2% of PWE had onchocercal skin lesions. Conclusion: People with epilepsy in onchocerciasis-endemic villages in Cameroon present with a wide clinical spectrum including nodding seizures and Nakalanga features. A great majority of participants met the diagnostic criteria for OAE, suggesting that better onchocerciasis control could prevent new cases. Epilepsy management algorithms in these areas must be adjusted to reflect the varied seizure types.